PrimeQA: The Prime Repository for QA
  • Installation
  • Development
  • Testing
  • API
  • Tutorials
  • GitHub
  • Slack
  • dense
    • colbert_top
      • colbert
        • data
        • distillation
        • evaluation
        • index
        • indexer
        • indexing
        • infra
        • modeling
        • parameters
        • ranking
        • run_indexer
        • run_searcher
        • run_trainer
        • search
        • searcher
        • trainer
        • training
        • utilities
        • utils
      • utility
        • evaluate
        • preprocess
        • rankings
        • supervision
        • utils
    • dpr_top
      • dataloader
        • distloader_base
      • dpr
        • biencoder_gcp
        • biencoder_hypers
        • biencoder_trainer
        • config
        • dataloader_biencoder
        • dpr_util
        • faiss_index
        • index_simple_corpus
        • searcher
        • simple_mmap_dataset
      • torch_util
        • hypers_base
        • transformer_optimize
      • util
        • args_help
        • line_corpus
        • reporting
  • run_bm25_retrieval
    • primeqa.ir.run_bm25_retrieval.get_language_id
    • primeqa.ir.run_bm25_retrieval.handle_args
    • primeqa.ir.run_bm25_retrieval.load_queries
    • primeqa.ir.run_bm25_retrieval.main
    • primeqa.ir.run_bm25_retrieval.run_search
    • primeqa.ir.run_bm25_retrieval.write_colbert_ranking_tsv
    • primeqa.ir.run_bm25_retrieval.write_search_results
    • primeqa.ir.run_bm25_retrieval.write_xorqa_json
  • run_ir
    • primeqa.ir.run_ir.main
    • primeqa.ir.run_ir.ProcessArguments
  • sparse
    • bm25_engine
      • primeqa.ir.sparse.bm25_engine.BM25Engine
    • config
      • primeqa.ir.sparse.config.BM25Config
      • primeqa.ir.sparse.config.IndexingArguments
      • primeqa.ir.sparse.config.SearchArguments
    • indexer
      • primeqa.ir.sparse.indexer.PyseriniIndexer
    • retriever
      • primeqa.ir.sparse.retriever.PyseriniRetriever
    • utils
      • primeqa.ir.sparse.utils.get_language_id
      • primeqa.ir.sparse.utils.load_queries
      • primeqa.ir.sparse.utils.write_colbert_ranking_tsv
      • primeqa.ir.sparse.utils.write_search_results
      • primeqa.ir.sparse.utils.write_xorqa_json
  • util
    • corpus_reader
      • primeqa.ir.util.corpus_reader.corpus_reader
      • primeqa.ir.util.corpus_reader.is_csv
      • primeqa.ir.util.corpus_reader.is_tsv
      • primeqa.ir.util.corpus_reader.list_corpus_files
      • primeqa.ir.util.corpus_reader.lookup_by_aliases
      • primeqa.ir.util.corpus_reader.DocumentCollection
      • primeqa.ir.util.corpus_reader.Passage
primeqa

preprocess#

primeqa.ir.dense.colbert_top.utility.preprocess.docs2passages

Divide a document collection into N-word/token passage spans (with wrap-around for last passage).